Acoustic and mechanical tapping methods for in vivo assessment of the interfaces of bone implants are experimentally compared using implant models made of metal and.glass. The former (AcT method) is based on the measurement of the frequency and amplitude of the vibration of a sample induced by a tapping needle. The latter (MeT method) is based on the measurement of the time during which a tapping rod is contacted with a sample. The following results were found. The vibration of a sample induced in the MeT test is very different from that induced in the AcT test; the former is much larger in the maximum amplitude and much lower in the main frequency than the latter, although both are of damped type. Nevertheless, a discrimination between hard and soft interfaces by each test is almost equally clear. It depends on load directions to some extent, reflecting a mechanical difference of the interface and/or the surrounding bone. The MeT test appears to be more sensitive than the AcT test for implant models surrounded by a thin compact bone. For implant models surrounded by a thick compact bone, however, the former is a little less sensitive than the latter.